Pith. sign in

REVIEW 1 cited by

Beyond Random Matrix Theory for Deep Networks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2006.07721 v2 pith:JZDA4FQP submitted 2020-06-13 stat.ML cs.LG

classification stat.MLcs.LG
keywords spectraltheoreticaldeepmatrixobservedrandomwignerdensities
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

We investigate whether the Wigner semi-circle and Marcenko-Pastur distributions, often used for deep neural network theoretical analysis, match empirically observed spectral densities. We find that even allowing for outliers, the observed spectral shapes strongly deviate from such theoretical predictions. This raises major questions about the usefulness of these models in deep learning. We further show that theoretical results, such as the layered nature of critical points, are strongly dependent on the use of the exact form of these limiting spectral densities. We consider two new classes of matrix ensembles; random Wigner/Wishart ensemble products and percolated Wigner/Wishart ensembles, both of which better match observed spectra. They also give large discrete spectral peaks at the origin, providing a theoretical explanation for the observation that various optima can be connected by one dimensional of low loss values. We further show that, in the case of a random matrix product, the weight of the discrete spectral component at $0$ depends on the ratio of the dimensions of the weight matrices.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Spectral distribution of sparse Gaussian Ensembles of Real Asymmetric Matrices

    cond-mat.dis-nn 2025-07 conditional novelty 4.0 of 10

    A complexity-parameter evolution equation is used to derive spectral densities for sparse real-asymmetric Gaussian ensembles, with numerical tests that fit several constants to the data.

Pith tools